The Journal of Experimental Biology 202, 2951–2959 (1999) 2951 Printed in Great Britain © The Company of Biologists Limited 1999 JEB2213 COLOUR VISION OF DOMESTIC CHICKS D. OSORIO1,*, M. VOROBYEV2 AND C. D. JONES1 1School of Biological Sciences, University of Sussex, Sussex BN1 9QG, UK and 2Institut für Neurobiologie, Freie Universität Berlin, Königin-Luise Strasse 28–30, Berlin 14195, Germany *e-mail: [email protected] Accepted 6 August; published on WWW 13 October 1999 Summary The colour vision of domestic chicks (Gallus gallus) was short-wavelength-sensitive receptors, one comparing the investigated by training them to small food containers outputs of medium- and long-wavelength receptors and a decorated with tilings of grey and coloured rectangles. third comparing of the outputs of short- and long- and/or Chicks learn to recognise the colour quickly and medium-wavelength receptors. Thus, the chicks have accurately. Chicks have four types of single-cone tetrachromatic colour vision. These experiments do not photoreceptor sensitive to ultraviolet, short-, medium- or exclude a role for the fifth cone type, double cones, but long-wavelength light. To establish how these receptors other evidence suggests that these cones serve luminance- are used for colour vision, stimuli were designed to be based tasks, such as motion detection, and not colour distinguished only by specific combinations of receptors. recognition. We infer (1) that all four single cones are used, and (2) that their outputs are encoded by at least three opponency Key words: bird, colour, vision, behaviour, Gallus gallus, chick, mechanisms: one comparing the outputs of ultraviolet- and chromatic coding, photoreceptor, cone, object recognition. Introduction Birds have five types of cone photoreceptor: four single helpful to distinguish the number of photoreceptor types from cones and a double cone (Bowmaker et al., 1997). By the number of degrees of freedom in behavioural use of colour. convention, the single cones are called long (L), medium (M), This number (i.e. the dimensionality of colour space) can be short (S) and ultraviolet (UV) wavelength-sensitive (Fig. 1A) estimated by colour mixture experiments. If m appropriately (long-wavelength-sensitive is sometimes abbreviated to chosen primary colours are necessary and sufficient to match ‘LWS’, medium to ‘MWS’, and so forth). Each contains a any spectrum, colour vision has m degrees of freedom. different photopigment, and the spectral sensitivities of L, M Mixture experiments establish the dimensionality of colour, and S cones are narrowed by a coloured oil droplet which filters but do not show how it is coded neurally. Generally, this is by incoming light (Partridge, 1989; Bowmaker et al., 1997). The chromatic and achromatic mechanisms. Chromatic coding fifth type, double cones (D), makes up approximately half of involves subtractive (opponent) interactions between receptor all cone photoreceptors. These have the same photopigment as signals, while achromatic coding is by additive interactions or by the L cones, but a different oil droplet filter gives them a one receptor type. Human colour vision has a single achromatic broader spectral tuning (Bowmaker et al., 1997). mechanism, which leaves the remaining two dimensions of colour to be encoded by chromatic mechanisms, called red–green Chromatic coding and yellow–blue (Jameson and Hurvich, 1955; Wyszecki and Colour identifies lights or objects by their intensity or Stiles, 1982; Lennie and D’Zmura, 1988). It is likely that other spectrum, while chromatic cues distinguish stimuli of differing animals use opponency, but there are few direct demonstrations spectral composition, irrespective of relative intensity. This in behaviour, except for the honeybee (Backhaus et al., 1987; requires a comparison of signals from photoreceptors of Brandt and Vorobyev, 1997). Indirect evidence for chromatic differing spectral sensitivity, typically by chromatic mechanisms comes from threshold spectral sensitivities, i.e. opponency. An eye with n cone (spectral receptor) types takes detection of monochromatic light added to a white adapting field. n samples of the spectrum, so lights are represented by a point The spectral sensitivities of several animals, including the pigeon in an n-dimensional receptor space. To use this retinal Columbia livia (Remy and Emmerton, 1989) and a passerine information fully, and so have n-chromatic vision, subsequent Leiothrix lutea (Maier, 1992), are explained by a model which neural coding must retain n degrees of freedom. Humans have postulates that colour is coded by chromatic mechanisms three photoreceptor types and trichromatic colour vision (i.e. a (Vorobyev and Osorio, 1998). However, alternatives are not three-dimensional perceptual space). For other animals, it is ruled out (Brandt and Vorobyev, 1997). 2952 D. OSORIO, M. VOROBYEV AND C. D. JONES A UV S M D L on at least three chromatic mechanisms driven by four single 1.0 cones, but not the double cone (Vorobyev and Osorio, 1998; Vorobyev et al., 1998). This implies that these birds are tetrachromats, but the conclusion depends upon the model’s assumption that colour thresholds are set by receptor noise in 0.5 chromatic mechanisms. Ultraviolet colour vision Relative sensitivity Relative 0 The ability of birds to see ultraviolet light has recently 300 400 500 600 700 provoked much interest, and it is clear that the ultraviolet cone 1.0 B Expt 2 signal is used for mate choice and for finding prey (Bennett et al., 1996, 1997; Burkhardt, 1996; Andersson and Amundsen, 1997; Church et al., 1998). Derimoglu and Maximov (1994) applied a conventional test for colour vision, showing that 0.5 some passerines can discriminate ultraviolet-reflecting stimuli from any shade of grey, while (Bennett et al., 1996) found, in Expt 3 Expt 1 mate choice, that removing the ultraviolet is not simply Relative quantal flux Relative equivalent to lowering the intensity. However, it remains 0 300 400 500 600 700 uncertain how the ultraviolet and other receptor signals are compared or combined. A comparison of UV and S cone Wavelength (nm) signals would be good for encoding spectral variation at short Fig. 1. (A) Spectral sensitivities of chicken photoreceptors in vivo, wavelengths. Alternatively, it would be ‘reasonable’ for birds modelled from data on photopigment, oil droplet and ocular media to sum UV and S cone outputs to give a trichromatic eye with spectral absorbance functions. See text for details. The sensitivities high sensitivity for short-wavelength light. This is (in part) of each cone type are normalised to their respective maxima. because the low intensity of short-wave illumination means (B) Spectral composition of illuminants used in experiments, that the UV–S chromatic signal is relatively noisy, in which normalised to their respective maxima. See also Table 1. Cones are case trichromacy may be favoured (van Hateren, 1993; classified as being sensitive to ultraviolet (UV), short-wavelength (S), medium-wavelength (M) or long-wavelength (L) light. D, Vorobyev et al., 1998). Goldfish do become trichromats at low double cones. intensities, although by dropping the L cone signal (perhaps because water absorbs red light most strongly; Neumeyer and Arnold, 1989). Tetrachromacy We describe here how domestic chicks use colour for finding Amniotes inherited four cone photopigments from fish food. Specific cone types and chromatic mechanisms were (Bowmaker, 1991; Hisatomi et al., 1994). Two pigments were isolated using a combination of restricted illumination spectra lost by mammals, but they are retained by some fish (e.g. (Table 1; Fig. 1) and selected object reflectances. The evidence goldfish Carassius auratus; Bowmaker, 1991), reptiles (e.g. is that all four single cone types and at least three separate turtle Pseudemys scripta; Goede and Kolb, 1994) and birds. chromatic opponency mechanisms are used. A role for the D All these animals have the potential for tetrachromacy. For cones is unlikely (Vorobyev and Osorio, 1998), but not ruled goldfish, the mixtures of monochromatic test lights required to out. match various monochromatic or white standards make a convincing case that their colour vision is tetrachromatic (Neumeyer, 1992). Materials and methods Tetrachromacy has not previously been demonstrated Animals in birds. Palacios et al. (1990) and Palacios and Varela Male chicks (ISA-Brown) were kept in pairs under standard (1992) trained pigeons (Columba livia) to discriminate conditions (McKenzie et al., 1998), with experiments starting monochromatic standards from mixtures of monochromatic 7 days after hatching. Water and food (chick crumbs) were test lights, which were then adjusted to give the best possible freely available, except that food was removed 120 min before match. Two monochromatic lights were needed to match an experimental session. standards of 590 nm or 600 nm, and also 450 nm. In the middle wavelengths, two lights could not be matched to a 520 nm Stimuli and viewing conditions standard, which implies that pigeons are at least trichromatic. Stimuli were printed on paper using a colour inkjet (Epson This makes a good case that pigeons discriminate colours from Stylus-Pro 1440 d.p.i.) and laminated with Sellotape. They 450 nm to 600 nm (although intensity cues are not ruled out), were made into open-ended cones 25 mm long, 7.5 mm in but does not demonstrate that they are tetrachromats. diameter, and with a 12 mm equilateral triangular tab at the Pigeon and Leiothrix lutea spectral sensitivities (see above) base that could be used as food containers. The patterns were are predicted by a model postulating that colour vision is based tessellations of 6 mm×2 mm rectangles, 30 % of which were Colour vision of domestic chicks 2953 Table 1. Relative quantal absorptions by single ⌠ photoreceptors viewing a spectrally flat surface under the Qi = Ri(λ)S(λ)I(λ)dλ , (1) ⌡ three illuminants used λ Cone type where λ denotes wavelength, Ri(λ) is the spectral sensitivity of Experiment receptor i, and S(λ) and I(λ) are reflectance and illumination number Illuminant UV S M L D spectra, respectively.